The Evolution of a MySQL Database, From a Single Instance to HA With Disaster Recovery
summary
Shows a concrete path from one MySQL 8 instance to multi‑site DR: enable gtids enabled and durable binary logging, stream binary logs off‑server for continuous point‑in‑time recovery, use mysql-shell Admin API and clone to provision replicas fast, deploy an InnoDB replica set for manual failover (minutes RTO, <1s RPO), then move to a three‑node InnoDB Cluster for automated failover and zero‑data‑loss within a site, with asynchronous cross‑region cluster‑sets for disaster recovery.
AI-written summary · about
chapters (13)
- 0:00Start
- 6:47Next Level (level up)
- 13:23Durable Binlogs
- 15:37Point-in-Time Recovery: how does it work?
- 18:01Improved Point-in-Time Recovery (PITR)
- 18:49Enabling GTIDS
- 22:49One single MySQL instance with daily backups and external binlogs archiving
- 28:46MySQL InnoDB ReplicaSet - examples
- 32:35MySQL Router
- 34:31MySQL InnoDB ReplicaSet - Manual Failover
- 41:01MySQL InnoDB Cluster: consistency levels
- 45:46Consistency: AFTER (WRITE)
- 46:39Consistency: BEFORE_AND_AFTER
related talks
57mThe Evolution of a MySQL Database System
24mComparing the High Availability of MySQL with Modern Distributed SQL Databases
35mMySQL HA, Recovery and Load Balancing
46mDatabase Reliability Engineering for MySQL
27mDistributed MySQL Architectures: Past, Present and Future
48mDemystifying MySQL Replication Crash Safety
14mThe State of the Art of MySQL Replication
40mImproving Observability, Reliability, and Security of Relational Database...